[SciPy-User] Error in constants documentation?
nicky van foreest
Thu Apr 8 13:23:26 CDT 2010
I once read that the Beagle (the Mars spacecraft ) crashed because
some engineer forgot to convert inches to metres, or the other way
around. Since people use python at nasa, it might even be an
improvement to remove the English units rather than to keep them in
On 8 April 2010 00:14, John Hassler <firstname.lastname@example.org> wrote:
> On 4/7/2010 2:40 PM, Charles R Harris wrote:
> On Wed, Apr 7, 2010 at 12:32 PM, David Goldsmith <email@example.com>
>> On Wed, Apr 7, 2010 at 10:52 AM, Charles R Harris
>> <firstname.lastname@example.org> wrote:
>>> On Wed, Apr 7, 2010 at 11:02 AM, Florian Lindner <email@example.com>
>>>> Am Dienstag, 6. April 2010 22:08:53 schrieb Arthur M. Greene:
>>>> > It would seem that there is some confusion, in the
>>>> > constants.html, between force and mass...
>>>> Beside the wrong unit which is fixed now I don't see any confusion.
>>>> > Strictly speaking, kg is a unit of mass, Newton a unit
>>>> > of force. Weight is force, not mass: A gold brick
>>>> > floating in interstellar space is weightless but still
>>>> > massive. Pounds and kilograms can be equated, but only
>>>> > in some specified gravitational field (like at the
>>>> > surface of the earth, where we usually weigh things).
>>>> This is true for pounds-force and kilograms. Pounds-mass and kilograms
>>>> could be equated in any context. Pound itself is ambigous.
>>>> > So mass is the more fundamental quantity, since it does
>>>> > not depend on gravity for its value. In Imperial units
>>>> > (feet, pounds) the unit of mass is the slug:
>>>> > http://en.wikipedia.org/wiki/Slug_(mass). This is
>>>> > absent from the constants page.
>>>> Mmmh.. never heard of it though I read quite some English language
>>>> aerospace engineering literature. However I'm using SI units. I think
>>>> pounds-mass is more widely used as a imperial unit of mass.
>>> I recall slug being used in amateur rocketry books 50 years ago or so.
>>> But SI units are definitely simpler.
>> OK, since Charles opened the door: what about taking the bold,
>> forward-looking step of not supporting "Imperial" units at all? (I say
>> "good riddance.")
> Well, what if someone wants to know what a slug is in SI units? It's not as
> if it is a big problem to support, and the more conversions the better,
> IMHO. The point of having these things down in code is that one doesn't have
> to go looking when the unit turns up somewhere.
> I think you're always going to have to (in effect) look up the units ... or
> at least understand what they are doing.
> Back when I was a productive member of society, I used to teach this stuff.
> English units of mass and force are very difficult, and here's why.
> Someone, back in the dim dark reaches of history, before even I can
> remember, decided that one pound mass (lbm) should WEIGH one pound force
> (lbf). Now, Newton said that F = M*A ... force = mass times acceleration.
> The natural unit for acceleration in English units is foot/sec**2, so we
> would naturally expect that:
> l lbf = 1 ft/sec**2 * 1 lbm. BUT .... weight is the force caused by the
> acceleration of gravity, and 1 lbm WEIGHS 1 lbf, so:
> 1 lbf = (32.174 ft/sec**2)*1 lbm.
> There seems to be a little problem here. (Metric units _define_ force in
> terms of mass and acceleration, so this difficulty doesn't arise. English
> units try to fix more variables than there are available degrees of
> freedom.) There are two alternative ways to fix it. One is to choose a
> more rational system of units. The other is to use a fudge factor. English
> engineers, naturally, chose the latter.
> The most common "fix" in Chemical Engineering is to define a fudge factor,
> the "gravitational constant," gc, such that:
> force (lbf) = acceleration (ft/sec**2)*mass(lbm)/gc. So gc = 32.174 ft
> lbm/(lbf sec**2)
> (Another "fix" is to use mass in terms of "slugs," where a "slug" is the
> mass which is accelerated by 1 ft/sec**2 by a force of 1 lbf. So a slug =
> 32.174 lbm. This doesn't seem to be used much any more.)
> The units problem is a real stumbling block for engineering students
> whenever we use English units (which is most of the time, in American
> industry). Some equations require gc, and some don't, and it isn't always
> obvious which is which. Sometimes "lbm" is hidden. For example, viscosity
> "contains" lbm ... except of course, if we use kinematic viscosity, which
> doesn't ... so equations containing viscosity need to have gc, except when
> they don't.
> And, of course, Murphy's law applies.
> SciPy-User mailing list
More information about the SciPy-User